After one year of Trump, Google CEO Sundar Pichai sounds a lot different about immigration — and that might be a …

Change in tactics but not substance. Whether it will work…

Almost exactly one year ago, Google CEO Sundar Pichai stood in front of cheering employees vowing to “stand together” and “never compromise”in opposing the Trump administration’s travel ban.

On Friday, Pichai spoke out again in defense of immigration and of the benefits that immigrants bring to the US. But the India-born CEO, himself the perfect embodiment of the cause, seemed to have ditched his firebrand approach in favor of a a more diplomatic tone.

“It’s really important that we don’t make it a tech-versus-the-rest-of-the-country issue,” Pichai said on-stage, during a Q&A event in San Francisco organized by MSNBC, when asked about immigration by hosts Kara Swisher and Ari Melber.

Many of the big immigration issues remain unresolved, hot-button topics, including the travel ban, the fate of the so-called Dreamers, and the controversial visa system US corporations use to hire skilled foreign workers. Indeed, the deadlock over immigration policy in Congress helped trigger the federal government shutdown that began on Saturday.

There’s a lot at stake for Google, which recruits engineering talent from all over the world and which counts some Dreamers — immigrants who were brought into the US illegally by their parents but now have work visas — among its ranks.

Still, after a year of the Trump presidency, Pichai appears to have adapted to the political climate his company now operates in. Instead of “never compromise,” Pichai stressed the need for Google to play a “constructive” role in the immigration debate.

“We are very open to constructively reforming the H-1B process,” Pichai said, referring to the visas that allow US companies like Google to hire foreign workers.

This softening in tone may not be the capitulation it appears to be though.

Sure, as a publicly-owned company Google has a responsibility to its shareholders to get on with business and it would be silly to expect Google to go to the mat on any issue that doesn’t directly affect its bottom line.

But Pichai’s real message seemed to be that Silicon Valley needs to be smarter to win this battle.

“It’s up to us as tech companies to make the case as to why immigration is good for the country, not just for tech companies,” he said. “I think we have to do that better.”

Silicon Valley’s greatest strength has always been its ability to sell the world on its vision of the future.

Steve Jobs and Bill Gates put a computer on every desk.

If Google can convince hundreds of millions of people to visit its website every day, then surely it should be able to convince citizens and politicians about the merits of a diverse society.

Pichai knows he has a good argument on his hand, and, like a good tech product, he just needs to figure out how to sell it.

Source: After one year of Trump, Google CEO Sundar Pichai sounds a lot different about immigration — and that might be a …

Spanish use is steady or dropping in US despite high Latino immigration

Second generation effect:

Hidden just beneath the surface of the ongoing heated debate about immigration in the United States lurks an often unspoken concern: language. Specifically, whether immigration from Spanish-speaking countries threatens the English language’s dominance.

Language and immigration have long been politically linked in the U.S. When Farmers Branch, Texas, passed an English-only “requirement” in 2006, then-Mayor Tim O’Hare justified it by saying that “we need to address illegal immigration in our city and we need to do it now.”

The Farmers Branch city council voted unanimously to drop the controversial ordinance last November, but 31 states and hundreds of towns in the United States still have local English-only or “official English” laws.

The perception that Latino immigration has led Spanish to sideline or even overtake English in the U.S. is widespread. After all, Spanish is the second most dominant language in the country, after English. It is spoken by 48.6 million people: 34.8 million Spanish-speakers age 5 and older of various national-origin backgrounds, 11 million undocumented Latin American immigrants and an estimated 2.8 million non-Latinos who use Spanish in the home.

Census data on U.S. demographic changes project that by 2060 the Latino population in the U.S. – the group most likely to speak Spanish – will grow 115 percent, to 119 million.

But these figures don’t tell the whole story. As a linguist, I have studied Spanish-English bilingualism in Texas, California, Florida and beyond, and I can attest that Spanish is not taking over the United States. Far from it: Political fearmongering notwithstanding, Spanish actually holds a rather tenuous position in the country.

From bilingual to monolingual

How can the Latino population be growing rapidly while Spanish-speaking remains stable? The answer lies in oft-overlooked peculiarities of census data and in the particular linguistic history of the United States.

If one looks only at immigration patterns over the past half-century, it is true that the U.S. has been gaining Spanish-speakers. From 1965 to 2015, roughly half of all immigration has come from Latin American countries. This trend added some 30 million people, most of whom came speaking Spanish, to the American populace.

But this is only half the story. While new immigrants bring Spanish with them, research shows that their children tend to become bilinguals who overwhelmingly prefer English. As a result, the same immigrants’ grandchildren likely speak English only.

Linguists call this phenomenon “the three-generation pattern.” In essence, it means that non-English languages in the U.S. are lost by or during the third generation.

We can see this pattern playing out in data from the Pew Hispanic Center. Surveys show that in 2000, 48 percent of Latino adults aged 50 to 68 spoke “only English” or “English very well,” and that 73 percent of Latino children aged 5 to 17 did.

By 2014, those numbers had jumped to 52 percent and 88 percent, respectively. In other words, the shift from Spanish to English is happening nationwide, both over time and between generations.

Why English dominates

Language shift is not a naturally occurring phenomenon. Rather, it is a consequence of cultural forces that pressure speakers to give up one language to get another. These forces include restrictive language laws that formally prohibit the use of Spanish in educational or government settings, as Farmers Branch, Texas, did for 11 years.

Schools also drive the three-generation pattern. Even though Latin American parents often speak to their U.S.-born children in Spanish, those children almost invariably attend English-only schools.

There, they learn that academic success is achieved in English. As a result, first-generation children expand their vocabularies and literacy practices in English, not in Spanish.

They may also encounter negative attitudes toward Spanish from teachers and peers. For example, in October 2017, a New Jersey high school teacher was caught on video reprimanding three students for speaking Spanish, encouraging them, instead, to speak “American.” That no such language exists is beside the point – her message was clear.

Social pressure to speak English is so great that Latino immigrant parents may notice resistance to using Spanish at home as early as kindergarten. A generation later, though grandparents may continue to use Spanish in the home, grandchildren will often respond to them in English.

The numerous blogs, websites and guides dedicated to helping Latino parents navigate this bilingual terrain indicate just how common language shift is.

Indeed, when I ask my own Latino students about when they speak what to whom, the answer is almost always the same: Spanish with elders, English with everyone else.

This pattern seems to hold in small towns and big cities, on the East Coast and on the West, and in towns with large and small Latino populations. From Chicago to Southern California, children of Spanish-speaking immigrants become English-dominant.

The Spanish-to-English shift even occurs in Miami, where over 65 percent of the population is Latino and where speaking Spanish has clear economic benefits. That’s why Miami struggles to find enough Spanish-speaking teachers to staff its public schools.

English on the rise

Spanish isn’t the only immigrant language that has struggled to keep a foothold in the U.S. Germans, Italians, Poles and Swedes went through similar language shifts in the 19th and 20th centuries. These languages, too, were sometimes seen as a threat to American identity in their time.

Then as now, American anxiety about the role of English in U.S. society was totally unfounded. In the roughly 150,000-year history of human language, there has never been a more secure tongue than English.

More people worldwide do speak Mandarin and Spanish as their first language. But with some 400 million first language speakers and more than 500 million adoptive English speakers, English has a global standing enjoyed by none of the roughly 6,000 other languages spoken worldwide. It has been that way for about half a century.

If Latino immigration declines markedly in the U.S., language shift may actually lead Spanish to disappear across America. English, on the other hand, isn’t going anywhere fast.

via Spanish use is steady or dropping in US despite high Latino immigration

Why policy-makers should care about behavioural science: Supriya Syal

As someone who likes ‘nudge’ and related behavioural approaches, liked this article for the examples it cited, both outside and inside Canada:

Classic examples of behavioural interventions for the public sector include measures that help people save more for retirement, consume less electricity, be more likely to pay taxes, sign up for organ donations and get jobs. But the application of behavioural science to policy-making is becoming increasingly diversified around the world. For instance, the Copenhagen Airport in Denmark, in partnership with iNudgeyou (a Denmark-based social purpose company), combatted the problem of people smoking just outside terminal entrances by using stickers that guide them to a designated smoking area a few metres away. Instead of telling people what not to do (no-smoking zones near doors), providing guidance on what behaviour is desirable (smoking zones away from doors) proved more effective. In South Africa, the government of the Western Cape, in partnership with Ideas42 (a US-based nonprofit behavioural design and consulting firm) and the University of Cape Town, used a computer-based “HIV risk game” to educate at-risk youth about HIV. Instead of the typical one-off information brochure, a gamification approach that got youth to make repeated decisions about HIV risk and provided immediate feedback about their choices was a more powerful aid for increasing young people’s understanding of such risks. In Australia, Alfred Health (a hospital), working with Deakin University, Monash University, VicHealth and the Behavioural Insights Team (UK), increased healthier food choices in its cafés by using a traffic-light colour system to classify nutritional value and portion sizes of beverages. Instead of telling people about the health costs of sugary beverages, the team marked drinks as red (most unhealthy), amber or green (most healthy), and “red” drinks were simply removed from the displays and self-service refrigerators and placed under the counters — still available for sale, just less obviously so. The total number of beverages that were sold didn’t change, but the sale of unhealthy sugary beverages went down by 28 to 71 percent in a range of trials.

Meanwhile, in Canada, the Canada Revenue Agency is successfully using behavioural nudges to get people to pay outstanding taxes and to file their taxes online. The Privy Council Office Impact and Innovation Unit (IIU) is using behavioural insights to help Statistics Canada increase its survey response rates, and to help the Department of National Defence recruit more women into the armed forces. The Innovation Lab at Employment and Social Development Canada has used behavioural insights to help people find jobs and has partnered with the IIU to increase uptake of the Canada Learning Bond among low-income families. And the foremost public sector behavioural insights group in Canada, the Ontario Behavioural Insights Unit, has succeeded in increasing organ donation rates and online licence-plate-sticker renewals using behavioural insights.

Despite these initiatives and widespread academic expertise, support and talent, the public sector application of behavioural science in Canada lags behind that of other comparable nations, and the opportunity to create better outcomes for Canadians through behavioural evidence-based policy is immense. Moreover, the bulk of behavioural-insights policy work in Canada amounts to what one might call repair jobs — it is as if we were retrofitting old buildings to make them work in the modern-day context — and behavioural science interventions are used mainly to improve implementation or compliance with a pre-existing policy. Which is important, no doubt, but we could also be building new structures that benefit from modern advancements and would be much better suited to their users’ needs. Indeed, behavioural science provides powerful tools that could help us get it right at the policy design and formation stages. And, in addition to using behavioural science to design better for the public, we could also be using these insights to design better for public servants. But there are few, if any, applications of behavioural science interventions to organizational decision-making within the Canadian government or the bodies that it regulates, so this is a significant area for exploration.

Some of our greatest challenges today are large-scale, complex problems of public policy, public perception and public action: climate change, vaccinations against infectious diseases, diversity and inclusion of historically marginalized groups, humanitarian crises, to name a few. Behavioural science promises empirically validated solutions that are derived from an understanding of the human beings that make up that “public.” And it rests its propositions on evidence in favour of what works to help people themselves make decisions that are better for them, so they can lead better lives. A pretty irresistible call to arms, wouldn’t you say?

via Why policy-makers should care about behavioural science

Census data says you’ll make a lot more than your immigrant parents, but your kids won’t make as much as you | Toronto Star

The chart above breaks out the visible minority population by generation. While Black Canadians and Japanese Canadians have the highest percentage of third generation, the actual numbers are small for 25-54 years olds: about 24,000 and 12,000 respectively. The numbers of the other groups are all under 5,000 (many under 1,000), save for Chinese Canadians at just under 9,000.

Given the relatively small size, it may be premature to make this conclusion regarding the overall prospects for the third generation:

Children of immigrants make a lot more money than their parents but their kids won’t make as much as them, the latest census shows.

While visible-minority immigrants tend to earn less than their white immigrant counterparts, their kids more than make up the income gap between the two groups and also outperform their white peers in the second generation, according to a report by the Association of Canadian Studies based on 2016 census data.

Part of the study, to be presented at a national conference in March on immigration and settlement policies, examines the ethnic differences in after-tax incomes across first, second and third generations of immigrants by ethnicity in the prime working age between 35 and 44.

For immigrants — white or non-white — that upward socioeconomic mobility based on earnings fizzled by the third generation when all groups, except for the Korean and Japanese, made significantly less money than their second-generation parents.

According to Jack Jedwab, the report’s author, visible-minority immigrants made an average of $38,065 a year, compared to $47,978 earned by white immigrants.

Overall, children of visible-minority immigrants made a 47 per cent leap in their average earnings above their parents, making $55,994 annually, surpassing their white second-generation peers, who made $54,174 annually or 13 per cent more than their own parents. (The white group also includes those who self-identified as Aboriginal, who makes up 6.1 per cent of the group.)

While all children of immigrants of colour did better than their parents, some communities fared better than others.

Second-generation South Asians made the most progress, earning an average of $62,671, up from $38,978 from their immigrant parents. Their Chinese peers, who had the highest average annual income of all groups at $65,398, made 50 per cent more than first-generation Chinese immigrants who made $43,085.

 

“The entire second generation enjoyed a higher mobility though some communities were faring better than others,” noted Jedwab, who teaches sociology and public affairs at Concordia University.

The higher socioeconomic attainment, he said, can be partially attributed to immigrant parents’ expectations on their children to make up for the sacrifice they made for the move and seize on the better opportunities Canada has to offer.

“Education is certainly a key explanation and I would suggest that the value that children of immigrants attach to higher education is greater than is the case for the grandchildren of immigrants,” said Jedwab.

via Census data says you’ll make a lot more than your immigrant parents, but your kids won’t make as much as you | Toronto Star

x

Reports urge more diversity in tech sector to bridge ‘digital divide’

Canada catches up to the Silicon Valley conversation on the lack of diversity in tech (to be fair, there has been discussion that led to these reports):

The tech and innovation sector needs to do a better job of recruiting visible minorities to boost Canada’s economic output, two forthcoming studies say.

On Wednesday, the MaRS Discovery District, a Toronto-based innovation hub, will release Talent Fuels Tech, a report that found most job seekers in the region are visible minorities and argues for the development of a sector-wide strategy to build and retain a more diverse work force.

And the Ontario Incubator Diversity Report, an independent study by non-profit advocacy group Innovate Inclusion, examined four prominent organizations with incubator and accelerator programs – including MaRS – finding leadership and mentorship teams lacked African-Canadian, Latin-Canadian and Indigenous leaders. This, the report finds, can create a “digital divide” in the province that holds members of those communities back from contributing to the innovation economy.

While the pair of studies highlight that innovation hubs tend to unfairly leave crucial demographics of Canadians behind, they also chart pathways to more equitable recruitment strategies for hiring underrepresented visible minorities.

“As much as we celebrate diversity in Canada, diversity even more so in technology often starts and stops with gender,” says Jessica Yamoah, executive director of Innovate Inclusion. “Diversity within an organization will bring you perspective and new ways of doing things. A lot of times, there’s not enough credit given to our immigrant community.”

Innovate Inclusion’s report will be published later this month. It applauds Ontario’s well-known incubators and accelerators for “rising to the challenge” of gender diversity, and invites them to take further steps to get more members of the three underrepresented minority groups into their leadership teams. Examining publicly available information about various leadership groups – including boards and executive teams – at MaRS and Ryerson University’s DMZ in Toronto, Kitchener-Waterloo’s Communitech and Invest Ottawa, the report found only five people of African, Latin or Indigenous descent among the dozens of leaders. (Invest Ottawa and DMZ told The Globe and Mail that some of the report’s figures are incorrect, but applaud its intent.)

Ms. Yamoah, who is Ghanaian-Canadian, says the report is not intended to scold the organizations, but rather to set a starting point from which they can improve inclusion, which would only serve to benefit everyone involved. “As we evolve in the innovation economy, certain communities are being excluded, creating a larger digital divide,” she says. “If the numbers aren’t established as starting point, how do we measure success?”

Canada loses when underrepresented communities aren’t included in the broader conversation, she says. “That can definitely be reflected in the bottom line. Look at how the African-Caribbean community affects buying decisions in popular culture. Instead of just being musicians and artists in front of the mic, why can’t they be the executives behind the deals, and the technology that’s used?”

The report’s suggestions include that governments provide more support to incubators and accelerators that strongly demonstrate diversity in their ranks, and to earmark funds and procurement opportunities for entrepreneurs of underrepresented backgrounds. Incubators, it suggests, should design strategies to build more diverse leadership teams and better target underrepresented groups.

MaRS’s report also sought ideas to better tap into tech talent in the Greater Toronto Area – and found the answer in seeking a more diverse range of candidates. “We need to build a pipeline of talent so companies can stop going through a closed loop of referrals,” says Lekan Olawoye, one of the report’s authors and the lead executive of Studio Y, a talent development and leadership program at MaRS.

The report surveyed nearly 600 technology workers, finding that 63 per cent of job seekers were visible minorities, while 56 per cent were born outside of Canada. “Either they’re looking for work because their current environment is not conducive to them, or they are overqualified for their current role and raising their hand and saying, ‘Hey, I have more skills to provide,’ ” Mr. Olawoye says.

The report recommends developing a “sector-wide diversity-and-inclusion working group” to develop a comprehensive system of talent attraction and retention for Toronto’s tech sector. And one major key unlocking this talent, Mr. Olawoye says, is by dismantling biases ingrained in hiring. One such bias is the need for cultural fit, which the report suggests is less important than hiring talent with a mindset for growth. “The best person might not be the person you like the most that you’ll take for a beer, but it’s the person who will help your company grow,” Mr. Olawoye says.

Ryerson’s DMZ has incorporated a diversity guidebook, TechGirls Canada and TWG’s Change Together, into its membership process. “I think it comes to no surprise that the Ontario ecosystem, in terms of diversity, is lacking – there’s a lot of work to be done,” says Abdullah Snobar, its executive director. Mr. Snobar underlines the importance of incorporating the value of diversity into incubators, rather than treating it as a series of boxes to check off. “We want to see it rooted into the culture of an organization,” he says.

Spokespersons for both Communitech and MaRS said they aim to seek diverse leadership, though they seek primarily to reflect the communities around them in attaining ethnic diversity. “To be diverse, our boards and our executive should mirror our population,” says Jodi Marner, Communitech’s head of diversity and talent initiatives, who suggested that Kitchener-Waterloo region does not have a large African-Canadian or Latin-Canadian communities to hire from. “I agree we’re not mirroring our population, but we need to understand our population better, and what it’s made up of.”

Ms. Yamoah warns that suggestions playing down the need for more African, Latin and Indigenous leaders are why the report was done in the first place. “This sentiment is concerning as it would never be expressed in the context of certain sports, entertainment, or the criminal justice system where these communities are often over-indexed,” she say

via Reports urge more diversity in tech sector to bridge ‘digital divide’ – The Globe and Mail

Quebec City Muslims alarmed by increasingly public displays of racism one year after mosque shooting

Sad:

Rachid Raffa is tired and bitter.

It’s been 43 years since he chose to settle in Quebec City after leaving Algeria. But as his encounters with racism become more commonplace, he’s come to feel less at home.

“When I came to this country in 1975 I got off at the wrong airport,” the 68-year-old said during a recent lunch break from his job at the provincial Ministry of Transport.

“I should have landed elsewhere in Canada.”

Raffa has been an active member of Quebec City’s Muslim community for decades. In the 1990s, he was president of the Islamic Cultural Centre, which later opened a mosque in the suburb of Sainte-Foy. He still prays there regularly.

More recently, he’s watched with disgust as mosques around the city are increasingly targeted by vandalism.

Anti-Muslim tracts were plastered over three prayer spaces in 2014. Some had their windows smashed the following year.

Raffa’s sense of dread deepened when, in June 2016, a pig’s head was dumped outside the Islamic Cultural Centre with the words ‘Bonne Appétit’ [sic] in a card.

“My bus goes by the mosque and I often told my wife ‘May God protect this place.’ But it happened,” he said.

On Jan. 29, 2017, moments after Sunday evening prayer ended, a gunman entered the nondescript building in Sainte-Foy.

Six men were killed that night, five others were injured. Seventeen children were left without fathers and the entire city was shaken to its core.

The response to the tragedy was swift. Thousands gathered the next day in the cold, holding candles and walking in silence, to honour the victims.

In the days that followed, politicians denounced all forms of hate speech and promised to safeguard the rights of all citizens.

But the light that emerged during the city’s darkest hour faded quickly.

CBC News spoke to dozens of community members in the weeks leading up to the one-year anniversary of the shooting. They described having to negotiate casual racism, outright Islamophobia and persistent fears for their safety.

Several who agreed to speak on the record refused to appear on camera or have their picture taken. They were concerned they would be targeted afterwards.

The social harmony promised by Quebec’s leaders after the shooting has failed to materialize.

In its place are acrimonious political debates over identity and religious accommodation, a surge in activity of far-right groups and a spike in the number of reported hate crimes.

“Everything that touches Muslims has become explosive. And we are fed up. I am fed up,” said Raffa.

“I am completely overwhelmed that this tragedy has led to the rise of racist rhetoric in the public sphere, to the complete indifference of Quebec’s elite.”

A climate of fear

Shortly after the shooting, Quebec City’s Muslim community resumed its long-standing effort to acquire a burial ground in or around the city.

The city’s first mosque dates from the late 1970s. But families had to travel to Laval, 260 kilometres away, to bury their dead.

They thought they had found a suitable location for the cemetery in Saint-Apollinaire, a town only 40 kilometres outside Quebec City. Even the local mayor was on board.

But a citizens group arose in opposition, and the cemetery project was quashed by a slim majority in a referendum.

Source: Quebec City Muslims alarmed by increasingly public displays of racism one year after mosque shooting

Computer Program That Calculates Prison Sentences Is Even More Racist Than Humans, Study Finds

Not surprising that computer programs and their algorithms can incorporate existing biases, as appears to be the case here:

A computer program used to calculate people’s risk of committing crimes is less accurate and more racist than random humans assigned to the same task, a new Dartmouth study finds.

Before they’re sentenced, people who commit crimes in some U.S. states are required to take a 137-question quiz. The questions, which range from queries about a person’s criminal history, to their parents’ substance use, to “do you feel discouraged at times?” are part of a software program called Correctional Offender Management Profiling for Alternative Sanctions, or COMPAS. Using a proprietary algorithm, COMPAS is meant to crunch the numbers on a person’s life, determine their risk for reoffending, and help a judge determine a sentence based on that risk assessment.

Rather than making objective decisions, COMPAS actually plays up racial biases in the criminal justice system, activists allege. And a study released last week from Dartmouth researchers found that random, untrained people on the internet could make more accurate predictions about a person’s criminal future than the expensive software could.

A privately held software, COMPAS’s algorithms are a trade secret. Its conclusions baffle some of the people it evaluates. Take Eric Loomis, a Michigan man arrested in 2013, who pled guilty to attempting to flee a police officer, and no contest to driving a vehicle without its owner’s permission.

While neither offense was violent, COMPAS assessed Loomis’s history and reported him as having “a high risk of violence, high risk of recidivism, high pretrial risk.” Loomis was sentenced to six years in prison based on the finding.

COMPAS came to its conclusion through its 137-question quiz, which asks questions about the person’s criminal history, family history, social life, and opinions. The questionnaire does not ask a person’s race. But the questions — including those about parents’ arrest history, neighborhood crime, and a person’s economic stability — appear unfavorably biased against black defendants, who are disproportionately impoverished or incarcerated in the U.S.

A 2016 ProPublica investigation analyzed the software’s results across 7,000 cases in Broward County, Florida, and found that COMPAS often overestimated a person’s risk for committing future crimes. These incorrect assessments nearly doubled among black defendants, who frequently received higher risk ratings than white defendants who had committed more serious crimes.

But COMPAS isn’t just frequently wrong, the new Dartmouth study found: random humans can do a better job, with less information.

The Dartmouth research group hired 462 participants through Mechanical Turk, a crowdsourcing platform. The participants, who had no background or training in criminal justice, were given a brief description of a real criminal’s age and sex, as well as the crime they committed and their previous criminal history. The person’s race was not given.

“Do you think this person will commit another crime within 2 years,” the researchers asked participants.

The untrained group correctly predicted whether a person would commit another crime with 68.2 percent accuracy for black defendants and 67.6 percent accuracy for white defendants. That’s slightly better than COMPAS, which reports 64.9 percent accuracy for black defendants and 65.7 percent accuracy for white defendants.

In a statement, COMPAS’s parent company Equivalent argued that the Dartmouth findings were actually good.

“Instead of being a criticism of the COMPAS assessment, [the study] actually adds to a growing number of independent studies that have confirmed that COMPAS achieves good predictability and matches the increasingly accepted AUC standard of 0.70 for well-designed risk assessment tools used in criminal justice,” Equivalent said in the statement.

What it didn’t add was that the humans who had slightly outperformed COMPAS were untrained — whereas COMPAS is a massively expensive and secretive program.

In 2015, Wisconsin signed a contract with COMPAS for $1,765,334, documents obtained by the Electronic Privacy Information Center reveal. The largest chunk of the cash — $776,475 — went to licensing and maintenance fees for the software company. By contrast, the Dartmouth researchers paid each study participant $1 for completing the task, and a $5 bonus if they answered correctly more than 65 percent of the time.

And for all that money, defendants still aren’t sure COMPAS is doing its job.

After COMPAS helped sentence him to six years in prison, Loomis attempted to overturn the ruling, claiming the ruling by algorithm violated his right to due process. The secretive nature of the software meant it could not be trusted, he claimed.

His bid failed last summer when the U.S. Supreme Court refused to take up his case, allowing the COMPAS-based sentence to remain.

Instead of throwing himself at the mercy of the court, Loomis was at the mercy of the machine.

He might have had better luck at the hands of random internet users.

Source: Computer Program That Calculates Prison Sentences Is Even More Racist Than Humans, Study Finds

Undocumented Irish Unexpectedly Caught In Trump’s Immigration Dragnet : NPR

Always interesting to see who gets caught when the net is cast so wide. While the Irish man caught is the focus of the story, the overall data is revealing:

The Trump administration has been aggressively deporting foreign nationals home around the globe, from Somalia to Slovakia. Though Mexicans, Central Americans and Haitians make up nine out of 10 people removed from the United States, year-end figures analyzed by NPR show that deportations to the rest of the world have jumped 24 percent.

Some are from formerly “recalcitrant” countries that used to reject U.S. deportees but have now agreed to take them home. These nations include Guinea, Cuba, Bangladesh, Iraq, Vietnam and Afghanistan. Moreover, agents with Immigration and Customs Enforcement, ICE, are arresting more immigrants in the interior of the U.S. who have overstayed their visas.

A case in point — the unauthorized Irish in Boston.

“It’s really indiscriminate. ICE, in their aggressive tactics of detention, are going after the Irish as much as they’re going after any other nationality,” says Ronnie Millar, director of the Irish International Immigrant Center in Boston.

Sitting in the visiting room of the Suffolk County House of Corrections, Dylan O’Riordan, 19, wears a lemon-yellow jail jumpsuit and a bewildered expression on his pale face.

“I was aware how with Trump immigration was going to get a lot harder, but I didn’t pay as much mind to it as I should have, which was my first mistake,” he says.

O’Riordan was born in Galway, Ireland. Both of his parents had lived in Massachusetts before he was born and already had green cards. They brought Dylan from Ireland to the Boston area in 2010 on a visitor’s visa when he was 12 years old. He overstayed his 90-day visa, and began living his life like any other American teenager, though he was unauthorized.

At 19, he had a child with his girlfriend, Brenna, then dropped out of high school and went to work for his uncle’s roofing company. About four months ago, he and Brenna were shopping at a mall when they got into an argument. “It was nothing at all,” he says. “Some woman called the cops, said I was abusing my girlfriend.”

O’Riordan was arrested for domestic assault and battery, but Brenna refused to file charges. The county chose not to prosecute. O’Riordan had no prior criminal record, so the judge let him go.

Immigrants who overstay their visas are at a unique disadvantage compared to immigrants who illegally cross the border. When they apply for their visa, they waive their right to an immigration hearing if they end up staying after their visa expires.

O’Riordan’s lawyer, Tony Marino, points out that his client was brought here when he was a child, but ICE won’t budge.

“Their position has been, well, he waived whatever rights he had when he came,” says Marino. “Twelve year olds don’t waive rights! I’ve never seen anything like it. I can’t wrap my head around it.”

The ICE office in Boston sent a statement to NPR: “Dylan O’Riordan … overstayed the terms of his admission by more than seven years. ICE deportation officers encountered him in Sept 2017 after he was arrested on local criminal charges. ICE served him with an administrative final order of removal.” He is scheduled to be put on a plane to Dublin later this week.

“You look American, you sound American.”

Dylan O’Riordan is not an isolated case. Irish visa overstayers have been swept up in the administration’s nationwide immigration dragnet. Under strict new rules, anyone here illegally is a target — whether they’re convicted of a crime or not. In 2017, ICE deported 34 undocumented Irish, up from 26 the year before. The numbers are tiny compared to the 128,765 Mexicans ejected from the country last year, but in Boston’s closeknit Irish community the wave of arrests is big news.

via Undocumented Irish Unexpectedly Caught In Trump’s Immigration Dragnet : NPR

Sanctuary cities: Why it’s harder for Trumpism to take root in urban Canada – Adams and Norris

 Adams and Norris on how greater urbanization in Canada provides a degree of resilience to Trump-style politics:

In early 2007, former New York mayor Rudolph Giuliani led the pack of would-be Republican nominees for president, but some worried he was “too metropolitan” for heartland voters. On Saturday, another famous New Yorker, Donald Trump, marks his first year in the White House. Paradoxically, the Manhattan magnate’s supporters are overwhelmingly rural and small-town folks.

Big U.S. cities such as New York and Los Angeles – and even smaller places such as Miami and Dallas – loom large in imaginations far beyond America’s borders. As for Canada, we suspect most people around the world tend to imagine the country as defined more by wilderness than urban life.

Despite the lower profile of Canadian cities, however, they arguably exert more pull in the country’s political life than U.S. cities do south of the border. American cities are culturally potent but politically constrained.

One reason is that a greater share of Canada’s population is clustered in a smaller number of cities. America’s 10 largest cities contain just 8 per cent of the country’s population. The proportion of Canadians who live in Canada’s 10 largest: 31 per cent. That clustering in a relatively small number of places is even more evident when we include the suburbs. If we look at the census metropolitan areas of the top 10 Canadian and U.S. cities, we find about a quarter of Americans (27 per cent) and more than half of Canadians (55 per cent) living there.

But it’s not just the fact of urban living that matters; it’s also the nature of the cities. Canadian cities are some of the most diverse on Earth. The populations of two of its largest, Toronto and Vancouver, are almost half foreign-born and more than two-thirds first– or second-generation Canadian. Our cities are largely products of postwar immigration. The past half-century has been especially important: Canada retired its explicitly racist immigration policies in the 1960s, moving to a points system prizing education and language proficiency, leading to huge inflows of talent, energy and youth from around the world.

The United States also had considerable (but proportionally smaller) migration inflows over the same period, which affected cities profoundly. But U.S. cities were also being shaped by forces related to slavery and segregation. In what’s called the Great Migration, millions of black Americans fleeing the violence and oppression of the Jim Crow South moved to northern cities such as New York, Milwaukee, Detroit and Chicago. In many urban neighbourhoods, as black residents moved in, whites moved out to monocultural suburbs – a pattern sometimes called “white flight.” Redlining – denying services to residents of certain areas – housing discrimination and other racist practices also contributed to the de facto segregation of ostensibly integrated cities. The effects of these policies remain to this day.

It’s true that poverty is racialized in Canada and that this is reflected in some of the residential patterns we see in and around big cities. But Canada never had a demographic upheaval on the scale of the Great Migration, which saw the internal movement of about six million Americans. The story of ethnic concentration in Canada is a nuanced one, shaped directly by discrimination in some cases – and indirectly by economic circumstances born of discrimination – but also often driven by people choosing to be close to others of their own background. Ethnic enclaves can support shops with offerings from “home,” as well as community and religious gathering places. The thriving Chinese community in the affluent Toronto suburb of Markham and the South Asian community in Surrey, B.C., for instance, were formed more by affinity than discrimination (which is not to say their residents don’t experience discrimination – just that it didn’t compel them to live where they live).

Destiny and geography

Another quality that differentiates Canadian cities from American ones is that they are connected to a system – and, importantly, a culture – of economic equalization. Although provinces are responsible for health and education, the federal government redistributes resources with the aim of ensuring that all Canadians enjoy comparable levels of service. This ideology shapes the political culture of provinces and cities as well; when disparities are revealed in the levels of service available to people living in different parts of a larger jurisdiction, Canadians tend to agree – at least in principle – that this is unacceptable.

Americans, with their greater skepticism of government and their greater attachment to local control, are less likely to believe that all Chicagoans, for instance, should enjoy the same quality of services. The fact that excellent schools funded by a strong tax base can be just a few miles away from struggling schools with crumbling infrastructure probably doesn’t thrill most Americans, but it is part of their economic and political tradition. Politically viable responses to such inequity (school vouchers, innovative charter schools) tend to be rooted in more individual choice and more entrepreneurialism, not more redistribution of resources and greater social solidarity across social and geographic boundaries.

The composition and characteristics of each society’s cities have important political implications. In Canada, it’s difficult to win a federal election without winning over immigrants and their children, a powerful presence in many urban and suburban ridings. In the United States, for presidential candidates, the diverse urban vote is useful but not make-or-break. Equally important, the urban vote isn’t always diverse; it can be monocultural. Redrawing electoral boundaries can allow candidates to ignore certain people and still win. North Carolina’s lawmakers have twice been ordered by judicial panels to redraw that state’s electoral map because of extreme gerrymandering – one according to voters’ partisan affiliations, another by race.

As for the U.S. Congress, the composition of the House of Representatives, like our House of Commons, largely reflects the distribution of the population. But the U.S. Senate – much more powerful than our largely advisory upper chamber dedicated to sober second thought – gives hugely disproportionate powers to rural states: Wyoming (population: 585,501) has the same number of senators as California (population: 39.25 million). Indeed, the 26 least populous states, whose 52 senators constitute the majority, represent less than a fifth of the country’s population.

When all these factors are combined, they result in a Canadian political landscape where cities matter enormously and an American political landscape in which it’s possible for national political actors to work around cities.

Canada has racists and racism, and like elsewhere, some of them are feeling emboldened by recent political events. But the mechanics of our political institutions are such that, at the national level, courting the dominant-culture majority at the expense of smaller ethnic or religious groups is a dangerous game, as the Conservatives learned in 2015. In the United States, it can be a winner.

Many factors differentiate Canada from the United States. Our history, our institutions, our values, our public policies are all distinct. The fact that so many of us live so close together in a small number of diverse – in a few cases hyper-diverse – cities is one of the key factors that makes a politically dominant Trump-style backlash on a national scale in this country unlikely.

via Sanctuary cities: Why it’s harder for Trumpism to take root in urban Canada – The Globe and Mail

Canada’s immigration lab: What the country can learn from PEI

Good long and balanced profile in the Globe on PEI’s immigration strategy, with many common characteristics elsewhere in Atlantic Canada. Retention will remain a challenge:

On a crisp, late-autumn morning, locals shuffle through the Charlottetown Farmers’ Market, scooping up farm produce from the fertile countryside just a few minutes up the road, cuts of meat from local ranchers, fresh Prince Edward Island oysters (shucked while you wait) and an array of handmade crafts and preserves. The scene is familiar to anyone who’s visited the market during its 34-year history.

But as you head deeper in, the lilt of rural PEI accents is interspersed with less familiar sounds: Mandarin, Hindi, Arabic. A long line of market-goers waits at a stall serving foods from Africa. Young students carry on an animated conversation in Spanish in front of a Mexican food vendor. Other stalls serve Chinese, Indian and Middle Eastern fare.

“It’s a different atmosphere altogether from what it was back 20 years ago,” said Ralph Younker of Younker’s Farm Fresh Produce. Mr. Younker has had a fruit-and-vegetable stall at the market for the past 28 years, and he has watched as a new wave of immigrants has transformed the popular Charlottetown attraction. “In the last two or three years, we’ve really noticed the influx.”

About 2,500 new immigrants landed in PEI in 2017; roughly 90 per cent of which have settled, for now, in Charlottetown. About the same number arrived in 2016. The province has also taken in about 800 non-permanent foreign residents in the past year .

While those numbers might not sound like a lot to people living in Canada’s big, immigrant-intensive hubs, such as Vancouver, Toronto and Montreal, they’re a very big deal in a place the size of Charlottetown (population 36,000). The annual immigrant influx is equivalent to roughly 6 per cent of the city’s population – which, on a per capita basis, dwarfs the intake of municipalities elsewhere in the country….

via Canada’s immigration lab: What the country can learn from PEI – The Globe and Mail